AutoBCS: Block-Based Image Compressive Sensing With Data-Driven Acquisition and Noniterative Reconstruction

Abstract

Block compressive sensing is a well-known signal acquisition and reconstruction paradigm with widespread application prospect of science, engineering and cybernetic systems. However, the state-of-the-art block-based image compressive sensing (BCS) generally suffer from two issues. The sparsifying domain and the sensing matrices widely used for image acquisition are not data-driven, thus ignoring both the features of the image and the relationship among sub-block images. Moreover, it requires to address high-dimensional optimization problem with extensive computational complexity for image reconstruction. In this paper, we provide a deep learning strategy for BCS, called AutoBCS, which takes into account the prior knowledge of image in the acquisition step and establishes a subsequent reconstruction model to obtain fast image reconstruction with low computational cost. More precisely, we present a learningbased sensing matrix (LSM) from training data so as to accomplish image acquisition, therefore capturing and preserving more image characteristics. In particular, the generated LSM is proved to satisfy the theoretical requirements, such as the socalled restricted isometry property. Additionally, we build a noniterative reconstruction network, which provides an end-to-end BCS reconstruction to eliminate blocking artifacts and maximize image reconstruction accuracy, in our AutoBCS architecture. Furthermore, we investigate comprehensive comparison studies with both traditional BCS approaches as well as newly-developing deep learning methods. Compared with these approaches, our AutoBCS framework can not only provide superior performance in both image quality metrics (SSIM and PSNR) and visual perception, but also automatically benefit reconstruction speed.

Simultaneous High-Resolution Blind Image Reconstruction and Perturbation Defense

“…In [14], a deep learning (DL) scheme was exploited for block-based image compressive sensing. In particular, image priors were used adaptively in the acquisition stage, and a reconstruction system was developed to achieve fast image reconstruction.…”

Deep Encrypted Traffic Detection: An Anomaly Detection Framework for Encryption Traffic Based on Parallel Automatic Feature Extraction

“…In their work, the feature set with prominent discrimination is selected from unencrypted transport layer security (TLS) handshake information, DNS response information related to the destination IP address in TLS fow, and header information of HTTP fow within the 5-minute window of the same IP source address, and the network trafc with malicious behavior is identifed from encrypted network trafc by the machine learning method. Inspired by efcient feature extraction capabilities of deep learning technology [13][14][15], Wei et al [16] used a one-dimensional convolutional neural network (1D-CNN) to better ft encryption trafc data based on Anderson and other predecessors' work. In 2018, Yang et al [17] proposed two deep learning methods to classify encryption trafc.…”